President Barack Obama’s biracial identity has at various times presented political rivals with opportunities. Born to a white mother from Kansas and a black father from Kenya, Obama has a “blackness” that some suspect political opponents of having emphasized to exploit negative associations. Hillary Clinton, as his chief opponent in the 2008 Democratic presidential primaries, was accused by political bloggers of artificially darkening Obama’s picture in a television advertisement in an effort to win the Democratic Party’s nomination, although the Clinton campaign strongly denied having done this.
But a recent study by Chicago Booth associate professor Eugene Caruso, with Nicole Mead of Tilburg University in the Netherlands and Emily Balcetis of New York University, shows that it may not be necessary to manipulate a candidate’s physical appearance to try to sway voters. That’s because voters themselves will tend to see a mixed-race candidate’s skin color as lighter or darker depending on whether they think the candidate shares their political beliefs.
A person who belongs to a group tends to evaluate its other members favorably, especially when the person in question identifies strongly with that group. A political party is an example of one such group, and an affiliation with a party influences how party members react toward candidates who share, or do not share, their political beliefs. Political partisanship can even shape how people see a biracial candidate’s skin tone, their views likely reflecting society’s long history of associating the color white with good and black with bad.
In the first experiment conducted for the study, participants were asked to assess an unknown male biracial candidate for a government position. The researchers gathered multiple photos of the candidate and created lighter and darker versions of each photo. In some, the candidate’s skin tone appeared lighter than in the original, and in others it appeared darker. Participants were led to believe that the candidate either did or did not support their views. They were then shown lightened, darkened, and unaltered images of the candidate, and they were asked to rate how well each of the photos represented him.
Participants who generally agreed with the candidate’s political views consistently rated the lightened photos as the ones that best represented him. By contrast, participants who disagreed with his views reported that the darkened images were most representative of him. Moreover, those who said the lighter-skinned images were most representative were more likely to say they would vote for the candidate.
Skin Tone and Voting
Even a well-known politician such as Barack Obama, whose face and image is recognized globally, can appear slightly different in the eyes of people who have differing political beliefs. In the study’s second experiment, corresponding with the 2008 presidential race between Barack Obama and John McCain, Caruso, Mead, and Balcetis devised a similar scenario, but used photos of Obama rather than of an unknown candidate.
The researchers found that liberal-minded students were more likely to pick lightened images of Obama as most representative of him, while conservative-leaning students were more likely to say darkened photos looked most like Obama. For comparison, researchers ran the same experiment using lightened and darkened photos of McCain, Obama’s 2008 conservative rival. Because there is no ambiguity about McCain’s racial background, the researchers did not expect—nor did they find—any relationship between students’ political orientations and their perceptions of McCain’s skin tone.
Moreover, Caruso, Mead, and Balcetis found that the more participants believed that a lighter-skinned photo was representative of Obama, the more likely those participants were to say they intended to vote for him. Some participants leaned right while others leaned left, but the researchers found that even after taking participants’ political beliefs into account, participants’ preference for a lighter-skinned image of Obama continued to be a significant indicator of their intended vote.
“It’s somewhat surprising that a subtle difference in the perception of skin tone is related to such an important decision like who to vote for in the presidential election,” says Caruso.
The Role of Prejudice
It would be possible for the results of this study to be driven by a strong bias against individuals with dark skin, including black people. Participants who are prejudiced against blacks would likely find the darkened images to be most representative of candidates they dislike. Those participants would also be unlikely to vote for biracial candidates they perceive as black.
To control for such racial attitudes, Caruso, Mead, and Balcetis asked the students who participated in the study’s third and final experiment to complete two previously-designed tasks that measure implicit and explicit prejudice. The first task, known as the Black- White Implicit Association Test, measures whether and how much students associate blacks with negative concepts and whites with positive concepts. The second task, the Attitudes Toward Blacks Scale, measures how respondents feel about statements such as, “Generally, blacks are not as smart as whites.” Students filled out these surveys a week before the 2008 presidential elections and at the same time they rated Obama’s photos. Immediately after the elections, the students reported who they voted for.
Participants who had reported that a lightened photo of Obama was most representative of him were more likely to report that they had voted for Obama in the election, while participants who had said a darkened photo looked most like Obama were more likely to have voted for McCain. Even after taking racial attitudes into account, the results persisted—regardless of a participant’s racial bias, the preference for a lighter skinned Obama led to a stronger tendency to vote for him.
As for prejudice, the researchers found a statistically significant relationship between the Implicit Association Test and the photo ratings for conservatives, which suggests that the more the conservative participants associated blacks with unpleasant concepts, the more likely they were to rate darker photographs as more representative of Obama. However, liberals who displayed similar racial prejudice did not view the darkened photos in the same way. This result suggests that the conservatives’ racial bias reinforced their negative attitude towards Obama’s political beliefs and led them to consistently choose the darkened photos. By contrast, the liberals’ preference for Obama’s political views may have overridden their racial bias and led them to choose the lightened photos.
How a person sees a biracial candidate’s skin tone may also depend on the color of the beholder’s skin. Most of the students who participated in this study were white, and initial results from follow-up studies with black participants show slightly different results. Caruso and his colleagues find that black and white participants react similarly in the studies when the political system is seen as unstable, which is often the case during elections. However, when the political system is seen as stable, they find the relationship between political views and skin tone is reversed for black participants: they will tend to see candidates with whom they agree as darker instead of lighter.